Abstract: Memory deficits lead to distortion when long recall periods are used to assess alcohol consumption. We used the recently developed Internet-based cell phone-optimised assessment technique (ICAT) to describe the drinkingpatterns of youngpeopleover the course of Thursday, Friday and Saturday evenings and to compare the amounts reported during the drinking sessions in question with those in the retrospective baseline assessment. Methods Using hyperlinks in text messages sent to their cell phones over five weeks, 183 young adults in French-speaking Switzerland completed a total of 8646 questionnaires at 8 pm, 9 pm, 10 pm, 11 pm, midnight and 11 am the next morning over 1441 evenings. Findings Participants consumed an average of three drinks on Thursday evenings, four on Friday evenings and five and a half on Saturday evenings. The multi-group and multi-level latent growth curves showed that while the difference was minimal at the beginning of the evening, consumption decreased over the course of the evening on Thursdays, remained about stable on Fridays and increased on Saturdays between both genders. The amounts indicated in the evening assessments were up to twice as high as those indicated retrospectively. Conclusions Using participants’ cell phones, ICAT appears to be a convenient method for collecting alcohol-related data throughout the evening. Due to the significant impact of eveningdrinkingpatterns on the total amount consumed and related consequences, it is important to prevent the average increase of drinking that is likely to occur on Saturday evenings among youngpeople.